Iron Mountain Incorporated (IRM) Earnings Call Transcript & Summary

We are ready to get started. I am pleased to welcome Iron Mountain, CEO, Bill Meaney; CFO, Barry Hytinen. Welcome to our conference. Welcome back, Barry.

And I think just to maybe kick things off, I'll ask if you could maybe clarify what you believe to be the biggest misconception about your core business and then talk a little bit about how unit economics are holding up in the face of -- what's going on in the paper business and what you view as the puts and takes?

Okay. Well, all right. So thanks for that. Let me start off with the core business. So I'd like to refer to it as the financial beast in the sense that we're blessed to have that as a core for a couple of different reasons. One is that it is a high 70% gross margin business. And you can see that we're able to -- mostly through price, be able to continue to drive mid- to high single-digit growth out of that. And because it is mature, that makes it even more cash generative because we don't have to put a lot of CapEx into it. Now a lot of people say, well, is that sustainable? And you continue to see in our supplemental that, that's basically flat to slightly up volumetrically and the rest is on price. And the reason why we're pretty comfortable about that, virtually every customer that we still have, is sending us new boxes of documents. Now albeit if you say, well, are some of them actually dropping in terms of the total number of boxes that we have, yes, but it's more what I call, sorry to be a bit wonky, although my kids are studying engineering and they say, that I'm past it, but it is a second derivative effect. In other words, what's happened is -- what's happening is that how can you be going down volumetrically with some customers, while they're still sending you boxes, it's just because they used to send, let's say, 5 new boxes every year. Now they might be sending us 3. And so when they get to that very stable 15 years, is what the average life of a box is, when that gets destroyed, there's just more being destroyed in that vintage, if you will, then there's coming in. So there are some customers for sure that there's a headwind volumetrically. But overall, when you kind of put it together, new customers, old customers, old customers that have already stabilized and what I would call the new norm, kind of steady but lower rates in terms of growth is that we're flat to slightly up. And of course, in some of the expanding geographies, that helps as well. Now why are people willing to spend more? We're approaching $0.40 probably on average per cube, we measure in cubes versus a few years ago, at least when I started the company, that was $0.25. Part of it is because of -- I'd like to say is we're not the lowest price, but we're the lowest cost option for people to store with us. And it's not just around the security or the seven 9s of reliability. We're supervised by the Boston Fed. We're a HIPAA-compliant company. Yes, people who are highly regulated are always going to seek Iron Mountain as kind of a safe refuge. But more and more, it's about the optional services that we can add. So if you think about it, people are willing to pay more to store a box of documents with us because with the flick of a switch or a send of an e-mail, is we can add other services, some of these we call Image on Demand, which was really a service that you could say it was there before, but with some of the new tools that we built 5 or 6 years ago when we were the AI/ML partner of the year with Google, if we image something, is we create metadata automatically without a human even looking at the document, we can create a rich set of metadata, which allows that customer to be able to actually not only just manage that from a regulatory or compliance standpoint, but allows them to extract more value and kind of transform their business. So there are things like that, their Smart Reveal, Clean Sweep. There's a number of new products that we've been able to add along the side of the core storage business or the traditional storage business that people highly value. So you see that, yes, we are able to get these kind of mid- to high single-digit price increases annually through with our customers because they see the value, and you can see that the retention rate of our customers has been rock steady as it was a number of years ago.

So you talked periodically about upselling and cross-selling. Your website does a very thorough job of talking about the life cycle of documents and storage and cloud and tying a lot of this together. And I'm just interested in some of the investments that you have yet to make to further refine those capabilities at the back end or at the front end? Or have you made most of those investments already?

Okay. So I would first and foremost, we've made most of those investments. So if you think about the company has been on a transformation. So say, 10, 12 years ago, we were a 10 -- our total addressable market was about $10 billion for our products and services. As we sit here today, it's about $150 billion. So what happened during that period of time is, I would say, starting in 2015, 2016, we started making investments in adjacent areas. So I mentioned the intelligent document processing, algorithms that we are able to build with the help of Google at the time. And now that runs on not just Google, but the AWS cloud as well. So that was one thing that we did back in, say, the 2015, 2016 time is closing near to your heart, Jon, is the data center build out. And I think that's when we started getting on your radar screen is we bought the IO data centers, which was really building a platform for our data center business. So I would say that was a big investment at the time. But now our investment in data centers is greenfield, sometimes a brownfield, but it's a greenfield continuous build-out. And then the more recent was the, I would say, the platform play we did with ITRenew in terms of the asset life cycle management business. I think it is fair to say that we'll continue to do bolt-ons, like we did with Regency, which is one of the best operators in North America in terms of the asset life cycle management. But those are businesses that we're buying at 7x, and you're able to synergize them quite quickly. So I think what's happened now, if you look at the company, which is really the transformation is what was a 2%, growth company back 10, 11 years ago, as we sit here today, this is a 10% to 12%, top line growth company that delivers 9% AFFO per share. And it's in a platform with, as you point out, cross-selling across the portfolio because the think, all those businesses share is 95% of the Fortune 1000 have been customers of ours for decades. So the cross-selling to us is a really important part of the glue. And we put that together is that as Mr. Buffett says, don't underestimate the power of compounding, right? I mean we don't see any slowdown in terms of being able to print, say, that 9% AFFO per share growth that we've been highlighting since our Investor Day a few years ago, and that translates to a 9% or better dividend per share growth over a very long period of time. And in fact, if anything we have a tailwind because as some of these faster-growing areas like digital data center, asset life cycle management become a bigger part of the portfolio, that just adds more fuel behind our back to continue that growth. So we think that we now have a financial model that's built on these pillars of the business that we've assembled that have that synergy in terms of cross-selling that gives us a pretty strong model.

Sticking to the core business and then we'll pivot into some other areas. Are there regions or customer verticals that you think can keep the growth trend going? Anything to highlight that you're noticing in the last period of time that could -- or where you can see better penetration or just better growth?

So we're in over 60 countries now. So I wouldn't say that geographically, we're saying, we need to add more pins on the map. That being said, there are a few always high points, right? We continue to see Asia. But more specifically, if we double-click on India. I mean, India has been a place that we've been focused on. I'm always a little bit worried when it gets faddish, right? And I would say, the last 3 or 4 years, it is just I'm old enough to remember when I lived in Hong Kong, everyone said, what's your China strategy? Now everyone is saying, what's your India strategy? But in fairness to Iron Mountain, since I came into the company, we've been focused on expanding in India and Mark Kidd, who runs our data center and asset life cycle management business, I was dragging him on trips and now he drags me on trips to India, going back 5 or 6 years ago to identify the right entry point because, Jon, as you know, I think you'll keep me honest, but I think about 6 or 7 years ago, the total IT critical load in India was less than 200 megawatts, right? And when you think about -- compare India to Northern Virginia, I mean, it's just -- and so we saw that and we said -- and they passed legislation that all that data or a lot of that data had to come -- that was being held offshore in places like Singapore had to come onshore because of privacy. So we've been focused on India for a long time. And I think India is probably geographically one of the places that we've invested. I personally go there 4 or 5 times a year. The only country that reports directly to me is India because we think that if you don't want a lot of big companies, they look at India and it's a rounding error, it's like 2% of their sales, is we think that, that underutilizes the power of that market. And unless it starts at the top of the house that you're providing that kind of attention because it is a hard market. I mean, it's a very, very, very competitive market, but it's a place where not only can you win, but you can develop products that you can reexport. So I think that is a big part. In terms of the verticals, it's very similar to what we've always been is that especially, I would say, the highly sensitive, highly regulated markets, which is kind of health care, financial services and governments around the globe. And now, obviously, with the hyperscaler, if you think all the hyperscale customers that we serve on the data center side because I consider them almost the same because their customers are also focused on these highly sensitive verticals in terms of highly regulated, very sensitive areas. So I would say the hyperscalers cut across those verticals. And as a result, they're a big part of our data center business. I don't know, Barry, if you want to...

Jon, I would add that from a core standpoint, the physical volume, India is also a major opportunity for us. We've already got a good sized business there. But as you and I have discussed before, that's a market that is still only now starting to embrace a level of outsourcing of paper records to partners like us. We're one of about three players in that market and of size. And there's a tremendous amount of paper in that economy, both from a standpoint of what's sitting with private enterprise as well as the governmental agencies. And we see that is clearly an area where we are getting incremental volume to your question. But frankly, and just to hitchhike on to Bill's comments there, we're growing in India across all of our major business units. So our digital business is growing in India. Obviously, data center is growing in India and our asset life cycle management is very small, but we aim to grow it very significantly there over time. So India is a big play. And there are quite a few other albeit smaller -- much smaller markets than India, where we see a similar phenomenon where outsourcing of records is in the early stages, let's say, whereas more of our mature markets, we had seen outsourcing of records decades ago, literally. And so that's following a good pattern. I would say that the other thing as it relates to physical volume that's helping us is our commercial teams continue to win incremental business from our existing clients. As we get further penetrated, even with our largest clients, there's more volume to win. They might be storing volume that -- with another partner in one geography, we can commercially win that and/or as clients are going through a lot of commercial real estate changes in light of post-COVID employment levels, we do see some marginal benefit from clients sort of using their office space more efficiently and going in there and saying, okay, what can be thrown out, what needs to be securely destroyed where we can help them with that, and what needs to be stored in a very safe long-term solution, which obviously we can help them with.

So pivoting into data centers and maybe starting with India. So Web Werks has a little bit less scale than others, but has been in business for quite some time. But what does the journey look forward like there in terms of capital allocation? And then maybe hit rest of Asia as well, and then we can...

Okay. So well, I think that we're -- for my introductory remarks, we're very bullish on the opportunity in India. I mean, India, I would say that you're right, it's relatively small, but there's nobody that's really massive right now in India. Some of our customers have made large deployments. And I think that will ramp over time quite quickly just as we start building the land bank and building the land bank is the key thing in India. And I think we're making good progress there, and I think we're making good progress there, and that was the main reason why we chose Web Werks as a partner because we thought they were the right partner for us to be able to partner and actually build that land bank. So we're comfortable with where we're progressing, but India, it's one of those things where the building the land is the key area. But I wouldn't say our focus in India is -- I mean, in Asia it's only India. We do like parts of Malaysia, we like, for instance, is, I think, an important part of the ecosystem. And there are other parts in Southeast Asia that we're continuing to keep an eye on. But I think if you look at overall in the Asia Pacific market, I think India, for sure, is the prize.

So if we look at your land bank overall, it's fairly concentrated in 4 or 5 markets. I think Madrid is another one. And is that something that we could see -- is it going to be a similar mix going forward? Or are there areas where you have relatively less land bank because you've been so successful, say, in leasing, that you would want to kind of re-up.

So I think that -- well, there's a couple of aspects of it, and I would divide it into 2 bits. We continue to look at our existing record centers in terms of areas for what I would call kind of more edge deployments, let's say, somewhere between 10 and 20 megawatts in terms of looking at re-purposing it. So we'll continue -- and we continue to see more interest in that. But those are -- but you're right to say when we're normally thinking about land bank and where Iron Mountain is really going is the large campuses, like we have in North Virginia, like we have in Phoenix, like we're building out in Madrid. Amsterdam to a certain degree. So I think you can expect us to continue to look for the right markets to do that. I think if you look at the markets that we've been successful in and we keep building our land bank like London, like Northern Virginia, like Phoenix is that we do continue to see that adding to it is just we've had the unique experience that as quickly as we're adding to it, is we're leasing it, right, in those markets. Those markets are attractive markets, as we all know. I think the other thing is we're looking for kind of what I would call the North -- the new Northern Virginia markets, right? I mean, when we first went to Prince William County, people would say to us, well, that's not Northern Virginia, right? Now I think we were one of the first movers into Prince William County or Manassas. And now everybody is there and now there's to going be another further expansion to the south of Prince William County. And so you can imagine that we're actively looking in those areas. But I think those markets will continue to be important. It's not our only focus, but for sure, we want to make sure that we continue for our customers to have capacity in those key markets.

And I'll give a little bit of context for folks that maybe don't know our data center story quite as well. Today, if we built out our entire platform of power-assured land, we would build out to about 920 megawatts with no additions. Now if I was -- when I was here with Jon last year, it was about [ 780 ] at this time. So we've been adding and you should expect us to continue to add to our land bank to build those points. Interestingly, we operate 265 megawatts of which we're above 97% leased. And we were under construction on 305 megawatts. But what's very interesting about our story is we are 96%, 97% pre-leased on all of that. So we're not building the spec here, right? We're building a contract. And when you think about the types of clients that pre-lease in this industry, it's a very short list. It's the largest cloud hyperscalers, the best credit tenants you could have in the world. And generally, we're signing leases with the hyperscalers of 10 to 15 years in duration with escalators. And I would say returns have been improving nicely over the last several years. If you go back about 3-or-so years ago on hyperscale, we were probably writing cash-on-cash unlevered returns in the 7s to 8s that kind of vicinity. But I'd say at this point, nobody in the industry is probably doing less than 10, if not 11, 12. And that creates a very nice flow of business because when you think about it over the next couple of quarters, just say, 3 years, we are going to more than double the amount of operating megawatts we have operating and they're already leased. So then as it relates to how much incremental land you should be planning for, what we're trying to do is keep up with our team's selling capacity. We would like to at least acquire power and land of, at least, as much as we sign each year. So last year, for example, we signed 124 megawatts. This last year, we started the year by saying we lease up 100 megawatts, we did 97 through the first half and increase the guidance to 130. So you should anticipate us trying to keep up with at least or go ahead of whatever we're signing because when we look out there and see the secular demand that's coming from the hyperscalers and the power needs, we really don't see any slowdown. I mean it's accelerating, Jon.

What would book-to-bill be for hyperscale that in your experience? I imagine that's kind of a lumpy metric, but any kind of ranges to think about?

Yes. And this is one of the other reasons why I like hyperscale is because we're pre-leasing and then, therefore, building right to the contract, whereas on a colo, brand-new spec colo, you're going to have a period to lease up. So in a hyperscale situation, if you have a powered land, our team has constructed in as quickly as 11 months. But depending upon the location, it may take a little longer. So I think you'd see in our supplemental anywhere between 1 and 2, 2.5 year. And that it's somewhat depends on if the hyperscaler maybe wants it in phases, in which case, that actually helps with the construction. So it might be 1 phase in 1 year, another phase in 1.5 years or in 2 years, that sort of thing. But generally speaking, it's been within about a 1- or 2-year increment.

Any changes in customer preferences and/or willingness around different deal structures, whether it's powered shell versus turnkey, things like cost variability going forward, if you want to maybe not all the risk. I'm just wondering, has that balance of power and the negotiation in the last couple of years shifted at all?

Well, I think that the -- well, for sure, the -- because of the constraints on power, right? And the insatiable demand for -- especially with AI, you say that -- and you see it in the pricing is the power has shifted to the data center providers. That being said, we look at this as a long game, so we try not to overexaggerate. We really do see especially the hyperscale players, which are accounting for a very -- the lion's share of our leasing activity as partners, so you don't want to overplay it. And at the same time, is to be responsive. So we have done powered shell, as you know, in our Virginia campus for a very important customer. And they've done subsequent leases on what I would call kind of a straight lease arrangement. But this was important in this deployment for them to have a powered shell and we wanted to work with them to provide that. I mean, the returns were fine for us. Could we have made maybe even higher returns if we had done it in a classical way, perhaps, but we also want to be responsive. I think the other part of it, I mean, Barry, you might want to comment is, we also -- many times, it works out to both of our advantages in terms of the way that we think about deploying capital and capital returns because you referenced that these are AAA customers, many times. So there are many times where it's a win-win. They might be looking for a slightly different arrangement, which allows us to effectively get higher returns on the capital that we deploy.

Yes. There are a couple of ways that we've been doing that, Jon. One is when you've got that kind of quality tenant for, say, 15-year lease, that's a fairly attractive situation for construction loans, for example. And so we're essentially in some ways, kind of borrowing at the tenant's balance sheet and credit rating. And therefore, we've seen some of the construction loans that we've done, for example, to be even below where I could borrow at for the parent. And then similarly, on like that quasi-power shells, we're doing more than a power shell there, but it is closer to a power shell than what we traditionally do. We were able -- that was another, again, great tenant, long-term lease, very good economics. And it was attractive to third-party capital. So in that case, we sold a 45% interest in that when we hadn't yet put a shovel in the ground, and we did that at a sub-4.5 cap. So what that enables to do is use third-party financing as well as the partner's capital to essentially cover all the build. So we won't put any dollars into the capital construction. In that way, we've essentially recycled our capital from going into that construction project where we already had a guaranteed return and putting it into further growth as we've been trying to keep up with the demand. So there's a few ways there to kind of continue to improve the return to Bill's point.

So in 2025, you've got 14% of your TCV coming up for renewal. And just wondering based on the demand signals you're getting from your customers, what are your preliminary expectations about customer retention, mark-to-market spreads?

Yes. So if you look at our lease renewals, it generally runs in that level pretty consistently every year because, as Bill mentioned, we started as a colo company. But what you'll see when you look at my lease exploration table is that, you've got a lot of megawatts that don't renew for quite a while, right? Because as you think about it. If you go back 4 or 5 years ago, we were mostly signing colo. But over the last 4 years, we've been principally toward hyperscale. I think last year, we might have signed 85%, 90%, it was hyperscale long duration. So you're seeing the renewals push out for us. In terms of what will renew next year and also this year and similar to my expectations for last year at the start of the year is, we expect to have relatively low churn. I think next year is probably going to be -- I will -- I'm not giving guidance for next year, but historically, and as long as I've been here, Jon, I think we've been saying something like in the vicinity of 1 to 2 a quarter, that kind of thing. We've been underrunning that for a while. Mark-to-market on renewals has been very good. It's been trending higher and higher. And I'd say, if you look on a pure triple net type of situation, both hyperscale has been rising and the colo pricing has been rising. So in light of how generally tight supply is out there, I would say we would expect those trends to continue.

ALM, Bill, you talked a little bit about it. You've guided towards $900 million by 2026 at your Analyst Day a while back. Current run rate is sub $400 million. Just wondering what sort of strategic steps or other organic factors that would enable you to get to that growth target?

Well, I think just the organic growth will give us a long run towards that, right, because we had a little bit of pricing compression, I guess, 18 months or so ago that we all saw. But if you just look at the overall growth rates in terms of both volume, I think in the last call, we highlighted both volumetrically as well as price -- component pricing is that we'll get a long ways to the target that we laid out in our Investor Day. I think in addition to that, I mean, Barry, you might want to comment is that we continue -- to say it's a pipeline that's probably overstanding, but we continue to see the opportunity to do very nice accretive acquisitions, which are more of, I would say, a roll-up nature. In other words, relatively small bolt-ons where paying like 7x or less before synergy.

Yes. There is a healthy number of -- in fact, it's a very fragmented industry in the asset life cycle management space. And for the most part, they're relatively small. The asset that we picked up recently Regency Technologies, which is a fantastic business and great team. We paid 7.5x on a trailing basis. We're getting a lot of very positive synergies out that deal. That was probably one of the larger assets that's out there. And so we continue to look and be very disciplined about where we are with respect to deals. However, what we find is bringing in that processing capability in-house across the globe is a very, I think, a very accretive strategy. And I would say asset life cycle management, I recognize we're running short on time, I just want to mention is one of the three big growth areas in our business. We have three distinct business units that now collectively are about 25% of the company's revenue. They are collectively growing in excess of 20% on a compound basis. So that's asset life cycle management at roughly $400 million that Jon mentioned. That's our data center business, which we spend a lot of time talking about that's growing 25%, 30% compound and got a lot of visibility for the future. That's about a $600 million revenue business at this point. And then there's our digital solutions business, which has been growing north of 20% compound for quite a few years. It is about going to be on a run rate of about $500 million of revenue at this stage. And as we said on our last quarterly call, we just had the best bookings quarter we've ever had in digital. So we think that as the team has been investing in digital solutions and an asset life cycle management and furthering our land bank and data center, that those three portions of our business can power a lot of growth going forward, supplementing what is a very strong core business that you started the discussion with, Jon.

So the ALM is asset and with all this AI demand from 1 year, 1.5 years ago, just now getting deployed there's a lot of equipment that might become end of life in a few years' time. And I just wonder, do you think further out in terms of ways to capture some of that?

Yes. I think the -- well, I think on that point is that was the thinking behind when we bought the ITRenew business because we had been in the, what I call the end user part of the business and a relatively small organic basis, which the Regency acquisition helps build that out even more so, and we continue to look at acquisitions to build that out. But the ITRenew specifically was important to your point, is because that really gave us both the capability and the platform to really have the discussion with the hyperscalers because that's where you see a big part of the refresh as they're putting in latest GPUs to drive their AI or their large language model capacity for their customers. So yes, we do see that, that's kind of creating -- so a few years ago, when we had -- we mentioned about the supply chain, the supply chain issues mainly affected that part of the business because what we saw is we saw a number of our hyperscale customers actually extending the life of their servers. Now with the new AI, and we see that actually coming down again because they need to refresh a lot of that equipment. And when they refresh that equipment is that gives us a new -- or a boost in the revenue in that business. So yes, we do think that's tailwinds over the next few years in the business.

We went a little bit over, but I want to thank you both for your time. Appreciate your attendance.

Thanks for having us.

Thank you very much.